Own Your AI Stack: a whitepaper for energy operators

The opportunity

AI spending in oil and gas is compounding faster than almost any other category of operational technology in the sector. Upstream applications, exploration, reservoir characterisation, drilling, production optimisation, account for the dominant share, reflecting where the highest-value, hardest-to-replicate workflows sit. This is a story about who owns the intelligence layer on top of the world’s most expensive industrial data.

The pull is no longer experimental. Devon Energy reported 25% productivity gains from AI drilling optimisation at CERAWeek 2025; ADNOC’s ENERGYai programme compressed seismic model-build times by 75% in its first year. Yet the IEA’s Energy and AI Special Report (2025) found the single biggest barrier to scaled adoption is not data or compute; it is the missing internal capability to run a working model as an operational service.

The technical landscape

A working model is roughly 15% of the journey. The other 85% is the engineering stack underneath it, six layers that turn a notebook into a workflow that ships inside the operator’s security perimeter.

A model that performs beautifully on a curated 200-well training set fails on the operator’s actual 4,700-well portfolio because the LAS headers are inconsistent, the raster logs were never digitised, and the SEG-Y was acquired across three decades by four different contractors. This is not a failure of machine learning. It is a failure of plumbing.

The vendor response, the one that has dominated the last five years, is to package the plumbing into a closed platform. These offerings solve a real problem, but they solve it in a way that ingests operator subsurface data to train shared platform models, converting hard-won reservoir intelligence into a vendor asset.

Our approach

EarthScan is an AI engineering company. We are not a model lab, not a services boutique, and not a platform reseller. We build and operate the full stack, six capability layers from HPC up to deployment, and we hand the keys back to the operator at scale.

The foundation underneath the modules is Hulde, our procedural foundation model from the Hominis family, released with open weights under a RAIL licence. Hulde is purpose-built for a property that matters more in subsurface than in any consumer domain: hallucination resistance. It prioritises correct execution of a procedural workflow over fluent narration of one.

Case examples

Operator A, a North Africa independent, ran a six-month POC with a tier-one vendor on an unconventional play. The technical results were validated. Eighteen months later, the model is not in production. The reason is mundane and instructive: no one owned the integration layer between the vendor’s model environment and the asset team’s existing workflow.

Operator B, a Gulf region independent, engaged EarthScan to deploy Well Log Detection across a 47-well evaluation portfolio. After eight weeks, the agentic workflow reduced evaluation time from roughly 14 hours per well to under 90 minutes, with geoscientist review retained for anomaly cases and final sign-off.

Implementation roadmap

We deliver in three phases: Discover, Pilot, Scale. Discover is short enough to fit inside a single budget cycle and produces an engineering artefact, not a strategy deck. Pilot is long enough to expose every layer of the stack to real operator data. Scale is paced to the operator’s risk appetite, not the vendor’s revenue model.

By month twelve, the operator owns the platform: the weights, the pipelines, the agents, the runbooks, and the team know-how to run and extend it independently. EarthScan transitions to a steady-state support role, or out entirely.

Risk and mitigation

No engagement is risk-free, and an honest whitepaper says so. The dominant risks, in roughly the order they appear, are data readiness, integration debt, change management, and over-scoping.

• Data readiness is the most common reason a Pilot slips. Our mitigation is the Discover phase itself: we surface the gap before signing a Pilot scope, and if the data is not ready, we say so out loud.
• Integration debt: pilots designed as parallel sandboxes rarely convert. We insist on integration into the asset team’s existing tools as a non-negotiable Pilot requirement.
• Change management: a module no geoscientist trusts is shelfware. We retain explicit review for anomaly cases and surface model confidence and provenance on every output.
• Over-scoping: a Pilot that tries to deploy three modules in twelve weeks delivers none. We scope to one module, one asset, one workflow.

Conclusion and next steps

The decision in front of operators is not whether to adopt AI; the market is moving from $7.2 billion to $17 billion in five years and that question is settled. The decision is whether the AI deployed on your asset is owned by you or by a platform vendor whose interest is maximising billable services on your subsurface data.

The next step is a four-week Discover engagement on a single asset. The output is a signed engineering brief: what is ready, what is not, what the highest-ROI first module is, and what the twelve-week Pilot looks like. If the readiness is not there, we will say so.